Onslaught Machine
Why have we chosen to talk about Onslaught just mere weeks before the beginning of Lorwyn previews? Uh, no reason. I mean what would be the purpose of talking about the first tribal block now? Did I say first? Well, yes, of course it's the first. No block had a tribal theme before Onslaught. And the word first does not necessarily by its nature mean that there is going to be a second. Although obviously some day we'll revisit the tribal theme. I mean it's a pretty popular theme and there's definitely a lot of meaty veins of unexplored design space left. If we were to approach such a theme again I know we'd have a number of new takes on it. I've been thinking about how to do tribal again since Onslaught block. But that doesn't... Okay, okay I can't take it anymore. Lorwyn has a tribal theme. You made me say it. Are you happy now?
Whew, that was a big load off my chest. Also, the new card type is planeswalker. (Yes, Tarmogoyf isn't a liar.) You heard it here first. What? The planeswalker minisite went live last Thursday? Oh. Never mind. Go check that out, then. Anyway, Lorwyn is chock full of cool stuff. But that's not why we're here today. Today is all about the first tribal set.
For our trip down memory lane, I've decided to pluck a bunch of cards out of the set and talk about whatever strikes my fancy. Yeah, most of it will be design-oriented, but not all of it. Enough of me explaining what the column is going to be about. Let's get to the part where it's about that.
I think this card's design started as a joke. But as often happens with jokes in design, the card made it to a playtest and people played with it. And surprise, surprise, this card was actually fun to play with. It fit the goblin's sacrifice theme and the humor was the kind that we do, in fact, do in the game. (One day I'll have to write a column about how I see humor in each color—red currently gets the lion's share.) The only mechanical strike against it was that red traditionally doesn't have a lot of small fliers but at 
for a 2/2 we figured you weren't paying for the efficient flying body.
Ancestor's Prophet Cycle (Ancestor's Prophet / Catapult Master / Gravespawn Sovereign / Skirk Fire Marshal / Supreme Inquisitor / Voice of the Woods)
In modern day design we have a good sense of where the next block is going thematically before we finish the block before it. This allows us to design with the future in mind. But back in the day, this wasn't common practice. What this meant was that often we'd make decisions that we would kick ourselves for later. The classic example of this was the decision in Odyssey block to not include many of the major races. Goblin, Elves and Merfolk were all replaced with lesser-used races (and in the case of Merfolk with a completely new race). This would prove very problematic when we decided to make the next block (yes, Onslaught) a tribal one.
Why do I bring this up with this cycle? Because this is another example of Odyssey not knowing that tribal was coming. In Odyssey, I came up with a design for a new type of tribal lord. Rather than simply enhancing all of its minions, these lords would allow the player to get function out of each creature of the proper race. The poster child of this cycle was Patron Wizard.
Patron Wizard turns each wizard you control into a Force Spike. I was very happy with how the cycle came out. But then we came to Onslaught and we needed new lord designs. I was bummed as I couldn't believe I had just designed exactly what we wanted except a year earlier when it had no relevance to the set it was put into. Luckily, someone (and I'd tell you who if I remember which means it was doubtful that it was me—odds are it was Mike Elliott) came up with the idea of taking the Patron Wizard cycle and pushing it. Rather than every creature getting a small effect, you tapped many creatures to get a single large effect. I was very happy with how their design turned out.
One last bit of trivia: this one of a tiny handful of six-card cycles in Magic. (I can't think of another one off the top of my head, but I know if I claim this is the only one, I'll get eighteen letters telling me the cycle(s) I forgot. Yeah, I'll still probably get the letters.)
One of the little games we play in design is one I call "Finish Out the Cycle." I'll explain how it works with a very concrete example. Alpha comes out and in it is are two cards called Control Magic and Steal Artifact. The first is an Aura that steals creatures and the second, well, read the title. Then in Tempest, I made the version of the card that stole enchantments. Along comes Onslaught and I realize that we've missed one. There are four permanent types after all (well, for now; number five is coming soon—hey, I have a card idea!). Time to Finish Out the Cycle. I'm pretty sure the playest name was Steal Land.
This card is an example of a classic mistake we make in R&D. We come up with some triggered effect and then because the effect seems so minor we choose to not put any mana on it. It's interesting to note that Lightning Rift (more on this card below) came from design without any mana cost when it triggered and development added one. But that was direct damage. Everyone knows that's good. Flickering, on the other hand, hadn't quite proven itself yet. Nowadays we are much more thorough about adding mana costs to repeatable triggered effects. We do print a few no-mana versions these days but gone are the times when those cards skated through development without any playtesting.
A lot of people think this card was designed top-down; that it was created to match the idea of a spell that turns attackers to gold. No, this was just excellent work from the creative team. The card started as a variant of the black enchantment No Mercy. In fact, its playtest name was A Little Mercy. The idea I was messing around with was to create an enchantment that played into white's "don't mess with me and my boys" flavor (that is, white can be aggressive on anything that attacks it first—think the Federation). Instead of killing the creatures like No Mercy, what if it just kept them from ever attacking again. Hit me and become a wall (old school defenders). That's where the card came from.
Another example of us not working out things at quite the right time is the change to the race/class model. Notice it started in Mirrodin, the first set after the tribal block. (On the plus side, it did give Lorwyn block yet another tool to differentiate itself from Onslaught.)
It's interesting to note though that the precursor to race/class showed up in Odyssey. When we created our new races—the Aven, the Cephalids and the Nantuko—I pushed the idea of having them all have two creature types, essentially one race and one class. The big difference from modern race/class was that each race was locked into one combination. Aven were Bird Soldiers, Cephalids were Cephalid Wizards and Nantuko were Insect Druids. As the set evolved there were a few changes. Only the Cephalids that seemed like Wizards stayed Wizards and a few Aven got Wizard added onto their type line, but the general idea stayed.
Which brings us to this card. Aven Brigadier was one of the earliest attempts to mess with race and class at the same time on the same card. I liked the idea that the card could be used in a Bird deck or Soldier deck but would really shine in an Aven (a.k.a. Bird Soldier) deck.
One of the A-list themes to vote on for next week's Topical Blend #3 was "Other games that influence Magic design." This card was first designed for a game other than Magic and then ripped off by me. What game? A little game called Mood Swings. Unless you're a faithful Making Magic reader with a good memory, you've probably never heard of the game. Why? Because it's never been published. Mood Swings is my attempt at making a more mass market trading card game. While I'm very proud of the design, various circumstances keep interfering with it getting published, so I've been constantly tweaking it for nine years.
Anyway, all the cards in Mood Swings are named after moods and emotions. One such card was called Curiosity. The original version of Curiosity basically said, "Look at the other player's hand." Now, in Mood Swings you only ever get one hand of cards (don't worry, it's a fast game) so looking at the other player's hand means that there is no hidden information for the person who gets hit with Curiosity.
The next piece of information you need to know is that my main playtester for Mood Swings is my wife Lora. We go out to lunch on the same day each week and we most often play Mood Swings. Lora and I have collectively played more Mood Swings than the entire rest of the planet. (I would love for that one day not to be true.)
As a quick aside, when we were naming my first daughter (Rachel for those of you that don't keep my personal trivia in a handy notebook), Lora and I were stuck on the middle name. I wanted Emily and she wanted Diana. We both liked the other's choice but clearly were leaning towards our own pick. I convinced Lora to settle the mater over Mood Swings. Normally over lunch we play best two out of three (like I said above it's a very fast game). The winner of each match won a letter. The first person to spell out their choice for the middle name won it. (Conveniently both names were five letters.)
Who won? I guess I'd let Rachel Emily Rosewater tell you herself, but she isn't up yet for a guest stint in my column. Yes, I won. Now, I'm sure many of you think I was taking advantage of my wife, "fooling" her into playing a game I created to make such a decision. What you don't know is that Lora led "DIAN" to "EM" at one point and I barely eked out a victory so I don't feel I took advantage of her. (Plus our individual win rate is close to 50/50.) Although I should point out that the choice of middle names for Adam and Sarah (our three-year-old twins) did not involve any gaming.
Back to the original story. I created Curiosity and played with it against Lora. She hated it. Now, she really likes Mood Swings (as if her playing it every week wasn't a sign), so having her hate a card this strongly piqued the designer within me. We talked about it and I realized that what she disliked so much was the fact that the card took a lot of the mystery out of the game. She liked knowing something but didn't want to know everything especially early in the game.
That's when I stumbled into the idea that you only showed a small number of cards for the opponent to choose from. Early game this meant that you could protect and keep secret key cards but that in late game it was essentially a Coercion. I made the change and the card played great. Then one day while Lora and I were playing, I drew Curiosity. As I looked it in my hand, I said to myself, "This would make an awesome Magic card."
And what set were we working on at the time? Onslaught.
For some reason, certain iconic cards keep calling out to the designers. Ball Lightning is one such card.
Crown Cycle (Crown of Ascension / Crown of Awe / Crown of Fury / Crown of Suspicion / Crown of Vigor)
One of R&D's ongoing quests is to find ways to offset the inherent card disadvantage of Auras. The crowns were designed to fulfill that role while also trying to make them feel at home in a tribal set. While I think they play well, my modern design sensibilities feel like there might be a way to accomplish this goal in a way that's a little cleaner.
One of the important skills in Magic card design is finding ways to make two halves of a card feel connected. Cruel Revival does a cleverer job than most on connecting its two halves. One effect targets a specific creature type and the other targets not-that-creature-type. Plus, each effect zone-shifts with the graveyard, one putting a card into the graveyard and another one taking one out. I'm just pointing this all out because I really like the design of this card. It's takes something that could so easily feel clunky and forced and makes everything feel as if it's meant to be together.
I just want to point out a popular tribal design trick used on this card. By making a creature that has an effect that affects its own creature type you make a card that can stand alone because it can always affect itself. You might see more of this trick in Lorwyn.
Sometimes design reinvents the wheel and sometimes we don't. Verduran Enchantress has always been pretty popular. This card was designed to allow an Enchantress deck to exist without creatures (green oddly enough has this quirky creatureless side that shows up from time to time). Also, I thought it was cute that additional copies of the card would trigger the earlier copies.
While I'm not 100% on this, I'm pretty sure that this is the first Mark Gottlieb designed card to see print. In design, the card was called Gottlieb's Bolt. There was a lot of debate about this card because its mechanic is very swingy. Interestingly, because the random factor was something that player's felt they had the ability to manipulate the card seemed to go over better with the section of the audience that normal badmouths random cards such as those with coin flips.
This card was designed by Richard Garfield for Odyssey, originally called Goblin Gattling Gun. But when I told Richard that it would have to be a Dwarf because there were no Goblins in Odyssey (man, it's interesting how dumb that sounds with knowledge of what was coming), he asked us to hold the card until such time we had Goblins again. Richard felt that the top-down design required a Goblin. During Onslaught development, the team found a spot for Richard's card.
Gustcloak creatures (Gustcloak Harrier / Gustcloak Runner / Gustcloak Savior / Gustcloak Sentinel / Gustcloak Skirmisher
If you want to improve your Magic card design skills, here's a task I recommend: teach some people to play Magic. Why? Because it gives you a great sense about what parts of the game matter to players on the most basic level. One of my discoveries from years of teaching beginners was that novice players are scared to attack (I'm talking in generalities here). This led me to create the enchantment Reconnaissance in Stronghold.
I thought the card would allow players who are normally scared to enter the red zone to attack without fear of messing up. The Gustcloak creatures were the next step. Instead of needing an enchantment to grant this to your creatures, just make some creatures that already have the ability.
One way to push a theme is to punish those that aren't using it.
Every once in a while we'll make a second version of a legendary creature. Usually when that happens it's because the character has gone through some radical change. Kamahl, for example, went from being a vicious pit fighter to being a sage leader. As such, it was felt that he needed a new card to show off his new softer side (meaning he turned from red to green).
Here's the card I turned in:
Kamahl, Land Lord
2GG
Creature - Druid Legend
1/6
All lands get +1/+1.
G: Target land is a 1/1 creature until end of turn. It's still a land.
I knew that Kamahl needed two abilities and I wanted the abilities to connect. Also, I wanted to reflect the new nature-loving Kamahl. Well, nature-loving, butt-kicking Kamahl (once a pit fighter, always a pit fighter). I was quite tickled by the "All lands get +1/+1." I was sad when it got removed.
The other little subtle design was that the red Kamahl in Odyssey was a 6/1 so I made the green version 1/6. The final card still kept the power and toughness adding up to seven but they upped his power to make him feel more aggressive.
Lightning Rift / Krosan Tusker
So we were looking for a final mechanic for Onslaught. We had tribal; we had morph. What we needed was what I call a "deck greasing" mechanic. That is, each set we try to include something that helps players smooth their mana draws (aka help make sure that players get land when they need it and avoid it when they don't). While morph helped a little (you could still play the creature even if you had color problems) it didn't fill the role we needed. The mantra that kept coming up was, "We need something like cycling." We messed around with a number of mechanics, but none had the right feel for the set. Finally, I made a bold suggestion: "How about cycling?'
You have to realize at the time that we hadn't ever brought back a keyword mechanic except to grant it evergreen status. My argument was that mechanics were a tool and we shouldn't be afraid to use them again. Yes, others replied, but we shouldn't bring back a mechanic unless we can do something new with it. "Okay," I replied, "how about cycling 
?"
It became apparent that if I was to sell the group on cycling that I had to demonstrate what were new areas we could explore with the mechanic. So I sat down and typed up several pages worth of cycling cards. Lightning Rift and Krosan Tusker (or at least the cards that would evolve into those two cards) were both on my list. The first demonstrated the idea of having cards that triggered off of cards being cycled and the latter was an example of cards that had an additional effect when cycled. Suffice to say that my cards did their job and cycling was added to the set.
Mistform Cards (Mistform Dreamer / Mistform Mask / Mistform Mutant / Mistform Shrieker / Mistform Skyreaver / Mistform Stalker / Mistform Wall)
I often talk about how sets can change greatly during the design process. Onslaught was one such set. It didn't start as a tribal set, at least not as its central theme. That isn't to say, though, that tribal was nonexistent. The Mistform creatures appeared in very early versions of the set. In fact, it was the interest in the Mistform creatures that started the tribal boll rolling. And rolling. And rolling. (What's the point of having a creature that can change its creature type if creature types don't matter?) So if you enjoy tribal, thank a Mistform creature.
I have fond memories of this card. You see, back in the day we had a Wizards Invitational modeled after the Magic Invitational where sixteen Wizards employees competed for the right to create a Magic Online avatar. Anyway, I let me readers build all of my decks and the deck I used in the Tribal format was centered around legendary creatures (back when "legend" was a creature type) and Patriarch's Bidding. Yes, my opponent got all of his creatures back. I didn't care. Mine were bigger. I just wanted to point out that some of my favorite cards don't come from designing them but playing them.
Here's one of the early cards that started laying down blue's Protection from Other Stuff that we've begun embracing more as part of blue's pie.
While most of the cards in a block are designed to follow a similar flow, I always like to make a card or two that swims against the current. Case in point is Riptide Shapeshifter. Most Onslaught block cards encourage you to play a deck of all the same creature type. Riptide Shapeshifter encourages the opposite, building a deck only using one copy of each creature type. So yes, this was done on purpose.
I like this design quite a bit. I think because it represents to me the kind of designs you get to when you're looking into fleshing out a theme as inspiration. Yes, this is an example of "restrictions breeding creativity" (for those of you playing the Rosewater Bingo game at home, any BINGOs yet?). I don't think I would have ever gotten to this card if I didn't start with "make an artifact that does something artifacts do but rewards you for playing a deck with a lot of creatures."
Thus far in the history of Invitational cards, this is the only art that we as a company felt bad enough about that we commissioned a promo version of the card to allow us a second chance at the picture.
The one other thing I want to say about this card is that I feel bad about how the environment kept the card from living up to its potential. See, Patron Wizard had come out the block before and R&D was worried that the wizard deck was going to be real annoying so they nerfed it. The only card left standing was Voidmage Prodigy as we try to push the Invitational cards. The problem was Voidmage Prodigy is only good if the environment had a bunch of good Wizards to play and the Odyssey/Onslaught Standard was not such an environment. Sorry, Kai.
Filling the Onslaught
That's all I got for today. I hope my little jaunt through Onslaught was fun.
Join me next week for Topical Blend #3.
Until then, may your tribe play nicely with itself.
Mark Rosewater